Although large language models (LLMs) has shown great performance on natural processing (NLP) in the financial domain, there are no publicly available tailtored LLMs, instruction tuning datasets, and evaluation benchmarks, which is critical for continually pushing forward open-source development of artificial intelligence (AI). This paper introduces PIXIU, a comprehensive framework including first LLM based fine-tuning LLaMA with data, data 136K samples to support fine-tuning, an benchmark 5...

Abstract A pure quantum state of N subsystems, each with d levels, is said to be k -uniform if all its reductions qudits are maximally mixed. Only the uniform states obtained from orthogonal arrays (OAs) considered throughout this work. The Hamming distances OAs specially applied theory information. By using difference schemes and partitions, we construct a series infinite classes irredundant (IrOAs), then answer open questions whether there exist 3-uniform qubits 2-uniform qutrits, ( >...

To compare the differences in floral composition and functions between two types of microbiota, ileal contents feces were collected from Sprague Dawley (SD) rats fed a conventional or specific-pathogen free (SPF) environment high-fat diet (HFD), V3–V4 region 16S ribosomal ribonucleic acid (rRNA) gene these was then amplified sequenced. Compared with feces, about 60% bacterial genera ileum exclusive, low abundance (operational taxonomic units (OTUs) < 1000). Of bacteria shared few highly...

Si- and Al-doped g-C 3 N 4 exhibit excellent activity selectivity in producing HCOOH CH from the CO 2 RR respectively, with catalytic performance comparable to that of noble-metal doped .

Orthogonal arrays have many connections to other combinatorial designs and are applied in coding theory, the statistical design of experiments, cryptography, various types software testing quality control. In this paper, we present some general methods find Hamming distances for saturated asymmetrical orthogonal (SAOAs) with strength 2. As applications our methods, SAOA parents size less than or equal 100 obtained. We also provide SAOAs constructed from difference schemes by expansive...

In recent years,research on Educational Data Mining (EDM) has developed rapidly. However, most researches focus data source issues, and ignore the importance of preprocessing mining algorithms. This paper studied EDM, with a special educational big Firstly, it analyzed relevant elements EDM introduces technology based requirements application. Then introduced common algorithms their applications, finally discussed development trend

Gd doped flower-like hierarchical BiOCl with highly exposed (110) facets and abundant oxygen vacancies exhibits a good photocatalytic potential.

k-uniform mixed states are a significant class of characterized by all k-party reduced being maximally mixed. Novel methodologies constructed for constructing with the highest possible purity. By using orthogonal partition arrays, series new $k$-uniform is derived. Consequently, an infinite number higher-dimensional states, including those purity, can be generated.

The total output value of the construction industry (TOVCI) reflects its own development level to a certain extent. An accurate prediction industry’s is beneficial government’s dynamic regulation. grey model widely used for simple calculation process and high accuracy. Based on TOVCI China from 2017 2020, this paper constructs an FGM (1,1) model, calculates r by simulated annealing algorithm, forecasts in next few years. At present, Particle Swarm Optimization algorithm (PSO) employed...

AbstractOrthogonal arrays (OAs) are widely applied in the statistical design of experiments. The Hamming distances OAs play an increasingly important role various areas; however, it is difficult to use distances, especially for tight and mixed orthogonal (MOAs). For example, two with same parameters may have distinct due their different structures constructions. In addition, studies on scant, exception a study indicating that symmetrical OA strength 2 has constant distance. this article, we...

This article studies the Hamming distances and association schemes of orthogonal arrays. We present a general method constructing mixed schematic saturated arrays strength two produced from symmetric by using expansive replacement method. As applications method, we obtain many new Some examples are given to illustrate this construction

In this work, we introduce notions of quantum frequency arrangements consisting squares, cubes, hypercubes and a notion orthogonality between them. We also propose mixed orthogonal array (QMOA). By using irredundant proposed by Goyeneche et al. can obtain k-uniform states heterogeneous systems from QMOAs. Furthermore, some examples are presented to illustrate our method.

Quantum combinatorial designs are gaining popularity in quantum information theory. Latin squares can be used to construct mutually unbiased maximally entangled bases and unitary error bases. Here we present a general method for constructing arrangements from irredundant orthogonal arrays. As an application of the method, many new obtained. We also find sufficient condition such that improved arrays [10] equivalent arrangements. further prove array produce uniform state.

Considered here are robust subgroup-classifier learning and testing in change-plane regressions with heavy-tailed errors, which can identify subgroups as a basis for making optimal recommendations individualized treatment. A new subgroup classifier is proposed by smoothing the indicator function, learned minimizing smoothed Huber loss. Nonasymptotic properties Bahadur representation of estimators established, grouping difference parameter baseline achieve sub-Gaussian tails. The hypothesis...

Human Pose Estimation (HPE) has been widely applied in autonomous systems such as self-driving cars. However, the potential risks of HPE to adversarial attacks have not received comparable attention with image classification or segmentation tasks. Existing works on robustness focus misleading an system provide wrong predictions that still indicate some human poses. In this paper, we study vulnerability disappearance attacks, where attacker aims subtly alter training process via backdoor...

Chicken is an important economic animal that encounter various stressors including high temperature, stocking density, bacterial infections and transportation, etc. affecting the poultry production with serious loss. To be in response to varied stimulus, hypothalamic-pituitary-adrenal (HPA) axis activated through controlling of synthesis secretion glucocorticoids (GCs). Present study characterized a novel gene C5H11ORF96, demonstrated significant upregulation after DEX injection chicken,...

For realizing the dynamic optimization of measuring instrument calibration interval, predicting history data by modeling. First improved moving average method is used to modeling and predict development trend parameters. On basis this, BP network compensate predicted residual sequence, so as get more accurate forecasts. Then MA - prediction model given optimize interval dynamically. The verified through experiments. results show that has higher precision better universality.

We consider availability data poisoning attacks, where an adversary aims to degrade the overall test accuracy of a machine learning model by crafting small perturbations its training data. Existing strategies can achieve attack goal but assume victim employ same method as what uses mount attack. In this paper, we argue that assumption is strong, since may choose any algorithm train long it some targeted performance on clean Empirically, observe large decrease in effectiveness prior attacks...